Why Jargon is Bad, and Some Modern Resources for RPG Theory

[Aldarc]

That's fair, and I agree! But that also ties into the "endless loop" that has been described by Torner, et al., in the hobby.

A. Amateur hobbyist declaring that there's a problem in TTRPGs.
B. This problem is caused by inconsistent desires/agendas/types of players.
C. Therefore, a new typology of players will be announced (almost always with some types being more equal than others, in the George Orwell sense).
D. Based on that typology, a theory (or theories) of TTRPGs and/or game design will bloom, under the concept that the system itself will enable/encourage/assist in certain types of play.
E. Rinse, repeat.

Having dealt with a fair amount of MBTI/Socionics and some other pop psychology, I tend to be skeptical of these sort of typologies. It's why we are still dealing with "Alpha Wolf" BS. With roleplaying games, I find most of the player typologies, including more academic ones, tend to fall short of providing satisfying insight about players in one way or another IMHO.

I think typologies of roleplaying games - game system "families" and common features they share - would be of greater value. Then layer the X Cultures of Gaming on top of that.

In addition, because of the hobbyist nature, the past lessons keep getting forgotten. We just keep re-inventing the wheel.

...as it was written in the The Elusive Shift by St. Peterson the Evangelist?

Having not read The Elusive Shift yet, I can't really comment in full. I have no doubt that the hobby has been dancing around a lot of the same key underlying issues from the beginning. Likewise, I don't doubt that a lot of early lessons have been forgotten. I am not entirely sure if we are just re-inventing the wheel, even if we are dancing around those same issues. If we are, it's not without good reason, IMHO. Our hobby was changing, is changing, and will change again in response to demographic changes in our hobby and our surrounding culture. And the far greater reach and influence that video games and its theories will undoubtedly make on our hobby cannot be underestimated.A lot of debate that was once in zines moved to Usenet, then to forums, then to Google+ and since elsewhere across the net (e.g., Discord, Reddit, etc.). However, I suspect that a lot has been recontextualized over time. Call and response to past movements and the surrounding culture isn't exactly a new thing outside of roleplaying games. It's a pretty descriptive phenomenon when it comes to artistic and cultural movements as well. I tend to think that falling back on the phraseology of "just keep re-inventing the wheel" does a great disservice to this call and response of our hobby to itself and the wider culture.

 

[pemerton]

With a scenario with nodes, the most likely railroading situation is a dm not allowing a scene to ‘fail.’ The PCs show up to interrogate a key witness, but either don’t get the clues they would need, or don’t understand the clues given, for them to get to the next ‘level’ of nodes (‘D’ in the linked article).

<snip>
I think it can be made more dynamic in various ways. For example, there could be a timer (‘what happens if the pcs do nothing?’). Or the dm can just figure out how the pcs actions in one node would change the entire gamestate.

Everything you describe here seems to me to reinforce my view that node-based design is a type of railroading.

We have the idea of key witness - and from the context it seems pretty clear that the GM is the one who decided that this witness is key. We have a clue the players need - and from the context it seems pretty clear that the GM is the one who has determined that the players need this clue. We have a timer that I take it from the context would be set up by the GM, implementing changes to the fiction unilaterally. And we have the possibility of the GM making unilateral changes to the fiction without reliance on a timer or similar device.

You are describing a GM-authored fiction which evolves in accordance with the GM's conception of it. The fact that it is not linear doesn't change that.

 

[Malmuria]

Everything you describe here seems to me to reinforce my view that node-based design is a type of railroading.

We have the idea of key witness - and from the context it seems pretty clear that the GM is the one who decided that this witness is key. We have a clue the players need - and from the context it seems pretty clear that the GM is the one who has determined that the players need this clue. We have a timer that I take it from the context would be set up by the GM, implementing changes to the fiction unilaterally. And we have the possibility of the GM making unilateral changes to the fiction without reliance on a timer or similar device.

You are describing a GM-authored fiction which evolves in accordance with the GM's conception of it. The fact that it is not linear doesn't change that.

Sure, but then I would need another term to describe a dm who doesn’t let a scenario fail, or let the players ‘skip’ nodes, or have the world react to the PCs dynamically.

edit: I see this structure best for a mystery scenario (like CoC where you have handouts related to each node) When playing dnd I usually just have an environment with various things going on in different locations.

 

[pemerton]

Sure, but then I would need another term to describe a dm who doesn’t let a scenario fail, or let the players ‘skip’ nodes, or have the world react to the PCs dynamically.

I wasn't sure what the original term was, in relation to which your new term would be another.

 

[niklinna]

Everything you describe here seems to me to reinforce my view that node-based design is a type of railroading.

<snip>

You are describing a GM-authored fiction which evolves in accordance with the GM's conception of it. The fact that it is not linear doesn't change that.

What about classic map-and-key dungeons that have no script, just corridors and doors and rooms? I mean, in some strict sense that's railroading in that PCs can't just teleport wherever they want. (Oh wait, in D&D they can, once they have the spell. But anyhow!)

What about NPCs you can discover and talk to, who have a relationship diagram (map)? You can talk to Victor, who knows Jennell and Ernie, which is how you might get introduced to or otherwise find out about them. Jennell knows about Ernie but has never met him. Ernie knows nothing about Jennell. Ernie does know Marian, though, but never tells anybody about that; he is in fact having an affair with her and meets her on the sly. Maybe if the PCs spy on Ernie (or Marian, assuming they learn about her some other way) they will find that out. And so on!

So now you've got me wondering about the difference between structure and railroad. Players can—or perhaps, be allowed to—presumably find creative ways around structure (teleport spells, Streetwise skill tests, whatever), but a railroad pointedly disallows that. Or rather, a scripted railroad disallows (explicity or implicitly), or a GM disallows or allows (possibly in defiance of a script). This gets into territory of who has authority over what in revealing/exploring/generating a story. Does that make sense?

 

[overgeeked]

With a scenario with nodes, the most likely railroading situation is a dm not allowing a scene to ‘fail.’ The PCs show up to interrogate a key witness, but either don’t get the clues they would need...

Locking clues behind dice rolls is monumentally terrible design. If the only way for the adventure to continue is to gain a clue, it's automatically gained. Simple as.

or don’t understand the clues given

That's on the players. But thankfully, they're never locked into only visiting the nodes and there's the Three Clue Rule. They can still go wherever they want. It's just that the clues for the mystery are at the nodes. The Alexandrian talks about this at length in the rest of the series you linked to.

for them to get to the next ‘level’ of nodes (‘D’ in the linked article). The other way it might end up a railroad is if they end up at the next part of the scenario but they can’t progress because reasons.

The thing about node-based design that's being left out is the Three Clue Rule. The short version is that for any conclusion the referee wants the PCs to make, there should be three clues pointing to that conclusion. You want the PCs to eventually go to location C, there should be three clues pointing to it. That's what the notation is below the nodes in the link you provided. The zero node (blue) gives you three clues, pointing to A, B, and C. Whichever node you go to also provides three clues, the other two middle nodes and D. The PCs can go wherever they want, even from A to B or A to C, or vice versa.

The referee is advised to create or move clues if the PCs go "off the map" to point them back to the nodes. This is what Hickman calls soft bumpers. Gently pointing the PCs back to the adventure without preventing them from going where they want. A hard bumper would be something like an impassible mountain or other impenetrable barrier meant to keep the PCs at the right locations.

I read Alexander’s thoughts on scenario design as being able to create and run a pre-written scenario while at the same time decreasingly the likelihood that the dm will feel the need to railroad the pcs to the next scene.

Justin's node-based design is basically what Hickman's been doing for decades, most famously in the Dragonlance modules. He calls it a closed matrix or narrative bumper pool. It's a good mix of linear and open-world design. There is an event that the PCs need to deal with, but instead of it being strictly linear or an impossible to manage perfectly open world, you use node-based design to get the best of both worlds. The freedom of choice akin to an open world but the structure of a more linear adventure.

I think it can be made more dynamic in various ways. For example, there could be a timer (‘what happens if the pcs do nothing?’). Or the dm can just figure out how the pcs actions in one node would change the entire gamestate.

Absolutely.

 

[niklinna]

I wasn't sure what the original term was, in relation to which your new term would be another.

I don't know the original term either, but for the things listed I've seen "GM Force" (with a capital F) used.

 

[pemerton]

What about classic map-and-key dungeons that have no script, just corridors and doors and rooms? I mean, in some strict sense that's railroading in that PCs can't just teleport wherever they want. (Oh wait, in D&D they can, once they have the spell. But anyhow!)

Just to make clear what page I'm on, which I think is the same page that you're on: I'm thinking White Plume Mountain, Tomb of Horrors, the Caves of Chaos and the like.

I don't see this as railroading, because there are no "dependencies between scenes", "key witnesses", "necessary clues", and in general the players set their own agenda. There is no GM-authored objective that the players have to pursue. (Torchbearer formalises this by getting players to set their goals; though it also pushes a bit towards GM curation by having the GM offer a list of goals!)

I see this as the classic paradigm of straight-up exploratory play, often with a side-helping of gamism (can we beat the dungeon?).

Going back to the six-room dungeon we were talking about upthread, that's perhaps not the most gripping dungeon of all time (though I am thinking of stealing it for Torchbearer!) but I don't see it as a railroad. The players have choices to make, including how to respond to the Bugbears and Goblins, and I don't see that the GM is the one dictating what they should be doing.

What about NPCs you can discover and talk to, who have a relationship diagram (map)? You can talk to Victor, who knows Jennell and Ernie, which is how you might get introduced to or otherwise find out about them. Jennell knows about Ernie but has never met him. Ernie knows nothing about Jennell. Ernie does know Marian, though, but never tells anybody about that; he is in fact having an affair with her and meets her on the sly. Maybe if the PCs spy on Ernie (or Marian, assuming they learn about her some other way) they will find that out. And so on!

I see these as borderline cases. In principle, it can be just like the dungeon: the PCs just poke around and explore and take what wins they can get. But because the "environment" of interpersonal interactions is so much more charged than the environment of "Trap in room 3 that Halflings are too light to trigger", I think it is pretty hard to treat the social "dungeon" as a thing to be explored in the same way.

In practice I think the sort of thing described is likely to involve a lot of players seeing the GM set out their social situation and then when the PCs intervene in it, the GM making decisions that reflect their conception of how the social dramas "should" play out.

So now you've got me wondering about the difference between structure and railroad. Players can—or perhaps, be allowed—presumably find creative ways around structure (teleport spells, Streetwise skill tests, whatever), but a railroad pointedly disallows that. Or rather, a scripted railroad disallows (explicity or implicitly), or a GM disallows (possibly in defiance of a script). This gets into territory of who has authority over what in revealing/exploring/generating a story. Does that make sense?

It makes some sense, although - and as per my other new thread - I think this focus on the geography of the fiction as a source of constraint could do with further scrutiny.

And I think we're absolutely in the territory of who has what sort of authority. For instance, in the node-based adventure, the GM has authority to decide what the PCs care about. Which is expressed via phrases like "key witness", "necessary clue", etc.

 

[Lanefan]

OK, but then what's interesting isn't the "linearity" but rather the dependencies. So let's have a term to describe that.

As I already posted, The Green Knight has no such dependencies.

I got nothin' when it comes to making up terms.

But, to add even more to the fun, there's two levels of these dependencies; the presence of one of which will make an adventure linear while the other will not.

The first type, the one that to some degree forces linearity, is when something Must Happen in an earlier scene in order that a later one can be accessed and-or played. A bland-and-boring example might be the trope where the PCs must find a key in one area in order to get through an otherwise-unopenable door in another; or when a scene with the butler must take place or else there is no possible way the PCs can safely access the mistress of the house with whom they need to speak. Kinda railroad-y, yes, but also quite common. These are the type of dependencies many of us either don't like at all or tolerate if used sparingly.

The second one, that doesn't force linearity, is when while all scenes are more or less equally accessible the play of some will be (in some cases greatly) affected by whether or not other scenes have already been played.

An example here might be where the PCs are trying to rescue a prisoner, did they find a stealthy way of avoiding all the guards (i.e. bypassing a variable number of scenes) en route to the prisoner, or did they throw stealth to the wind and just blow away the guards on their way in.* Put another way, the scene where they reach and rescue the prisoner is in theory going to happen at some point if the PCs are to succeed in their mission, but that scene will play very differently if it is among the first to be played during the adventure rather than among the last. Note too that this scene sequencing is almost entirely dependent on how the players/PCs choose to approach the mission.

This second type of dependency is IMO just fine, and sometimes almost falls under simple consequences: if you do A first it'll make doing B either easier or harder than if A had not been done (and might even make doing B either impossible or unnecessary).

* - georgraphy factors in here as well. If the prison is physically laid out in such a way that the PCs have no choice but to plow through lots of guards en route to the prisoner then it's going to be a linear adventure no matter what, with a pre-known number of scenes in it. But if the prison is laid out in such as way as to allow different approaches (or the PCs can bring resources to bear to allow different approaches e.g. Passwall spell, flight, etc.) then the adventure need not be linear at all and - using number of scenes as the measure - the adventure's length cannot be predicted ahead of time. Maybe they do wade in and encounter every possible scene and-or location and have 6 combat scenes then the actual rescue scene then another combat followed by a chase; or maybe there's only three scenes to the whole thing - stealthing in, rescuing the prisoner, and stealthing out again.

 

[Lanefan]

This remark tends to reinforce the impression I'm getting that "linear adventure" has two uses:

* To describe geography (maybe with an additional premise in the neighbourhood, along the lines of where a PC is determines what scene the GM frames);

* To describe interdependencies between scenes, such that one has to finish a certain way (or within certain parameters) [...] for the next to be framed.

Correct, except that where I added "[...]" in your quote, put the words "or take place at all".

There is often a lot of overlap between these two things - geography and scene interdependency - in that in many published adventures each room or area has a specific encounter tied to it, meaning that if you are being geographically forced to hit the areas in a specific sequence you'll by default also be forced to hit those encounters in the same specific sequence.